INDEX

Explanations

filler words/adverbs

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

was

-1.49

 wasn

-1.27

 were

-1.18

was

-1.18

wasn

-1.18

 weren

-1.09

Was

-1.07

had

-1.06

 wasnt

-1.06

 consisted

-1.05

POSITIVE LOGITS

 ostavi

0.61

are

0.58

 will

0.57

HasForeignKey

0.57

Filmografie

0.55

 kohta

0.55

ține

0.55

 Jurí

0.51

明年

0.50

dır

0.50

Activations Density 0.135%